Nickel-based dual catalyst systems effective in suppressing carbon dioxide formation during diethylene glycol amination
Disclosed is a process for simultaneously producing morpholine and GIGLYCOLAMINE.RTM. amine while suppressing CO.sub.2 formation comprising the steps of:reacting anhydrous diethylene glycol with ammonia in the presence of added hydrogen over a dual nickel/ruthenium catalyst system comprising a nickel-based catalyst plus a supported ruthenium catalyst, andisolating said morpholine compound and said DIGLYCOLAMINE.RTM. amine compound from the reaction products.
Latest Texaco Chemical Company Patents:
- Controlled epoxidation of propylene
- Continuous isobutylene-assisted aqueous extraction of methanol from methyl tertiary butyl ether
- On-step synthesis of methyl t-butyl ether from t-butanol using platinum/palladium modified .beta.-zeolite catalysts
- Use of supported chromium catalyst in the preparation of tertiary butyl alcohol from tertiary butyl hydroperoxide
- Use of supported palladium/platinum catalysts in the preparation of tertiary butyl alcohol from tertiary butyl hydroperoxide
This application is related to U.S. Ser. No. 07/543,017 filed June 25, 1990.
Field Of The InventionThis invention relates to the practice of making morpholine from diethylene glycol by an amination reaction. More particularly this invention relates to the use of a dual nickel/ruthenium catalyst system for the selective amination of anhydrous diethylene glycol (DEG) to morpholine plus DIGLYCOLAMINE.RTM. amine (DGA) wherein the coproduction of CO.sub.2 is significantly suppressed. Surprisingly, it has been found that only certain combinations of nickel/ruthenium dual catalyst systems are effective in suppressing CO.sub.2 formation during morpholine service.
BACKGROUND OF THE INVENTIONThe practice of making morpholine plus DIGLYCOLAMINE.RTM. amine from diethylene glycol and ammonia over an amination catalyst is well-known in the art (see, for example, U.S. Pat. No. 3,151,112 (1964). The reaction is generally run in an aqueous environment. The indication is that better productivity could be obtained using an anhydrous feed, since water acts as a diluent and takes up space in the reactor. Whether run under aqueous or anhydrous feed conditions, there are two principal gaseous by-products, methane and carbon dioxide. Under aqueous conditions most of the CO.sub.2 is absorbed into the H.sub.2 O. Under anhydrous feed conditions, even though water is a reaction by-product (e.g. see Equations 1 and 2), the CO.sub.2 will readily react with the excess ammonia to form ammonium bicarbonate and/or ammonium carbamate, both of which in the absence of adequate water diluent, will form solids which precipitate downstream of the amination reactor, in the heat exchanger, etc., causing additional unit maintenance costs and engineering problems. It would be extremely advantageous if it were possible to operate a morpholine/DIGLYCOLAMINE.RTM. process using anhydrous diethylene glycol/ammonia feed in such a fashion that CO.sub.2 by-product formation was lowered to below, or close to, threshold levels (i.e.<50 ppm). Such a condition might be achieved by converting said CO.sub.2 to methane (Equation 3), possibly by introducing a methanation catalyst into the process unit, downstream of the diethylene glycol amination step. Particularly useful would be a methanation catalyst that:
a) Lowers the CO.sub.2 concentration in the morpholine/DIGLYCOLAMINE.RTM. amine product effluent to threshold levels.
b) Is stable in the high pH, amine, environment of the morpholine/DIGLYCOLAMINE.RTM. product effluent.
c) Remains an active methanation catalyst in said environment for extended periods, particularly at the temperature of DEG amination (190.degree.-250.degree. C.).
d) Does not convert the desired morpholine and DIGLYCOLAMINE.RTM. to unwanted by-products, as a result of catalyzing secondary reactions in the crude product effluent. ##STR1##
There is art pertaining to hydrogenation/dehydrogenation catalysts. In U.S. Pat. No. 4,508,896 there is described a process for simultaneously producing a 2-(2-aminoalkoxy)alkanol compound and a morpholine compound by contacting oxydialkanol with ammonia in the presence of a hydrogenation/dehydrogenation catalyst. In that process a hydrogenation catalyst was used which comprised about 60-85 mole % nickel, 14-37 mole % copper and 1-6 mole % chromium calculated on an oxide-free basis. This work does not address the problem of a methanation catalyst, CO.sub.2 production and its effect in lowering productivity. In said U.S. Pat. No. 4,508,896 to Templeton the hydrogenation/dehydrogenation catalyst is operated at a temperature of about 190.degree. C. to 230.degree. C. and a pressure of about 700 psig to about 2200 psig.
No art has been found dealing with a provision for converting by-product CO.sub.2 to methane when said CO.sub.2 is present in an amination product effluent, particularly where the CO.sub.2 is present in a morpholine/DIGLYCOLAMINE.RTM. amine product mix resulting from the amination of diethylene glycol (DEG) with ammonia.
It is known that carbon dioxide can be converted into methane by reduction with four moles of hydrogen (Equation 3). This is called the "Sabatier reaction". The mechanism for the reaction is not clear and early researchers assumed the reaction occurred in stages, however Bardet and Trambouze reported in work in C. R. Hebd. Seances. Acad. Sci., Ser. C. 288 (1979), 101, that it appeared CO.sub.2 methanation had its own mechanism, distinct from CO methanation, and that CO.sub.2 methanation occurs faster and at lower temperatures than methanation of CO. Heterogeneous catalysts have proven best for this process, see "Carbon Dioxide Activation by Metal Complexes," by A. Behr, Chapter 4, p. 85 (1988).
The mechanism for hydrogenation of CO.sub.2 on nickel is discussed in an article titled, "Hydrogenation of CO.sub.2 on Group VIII Metals II. Kinetics and Mechanism of CO.sub.2 Hydrogenation on Nickel," G. D. Weatherbee, et al., J. Catal., 77, 460-472 (1982). The rate of CO.sub.2 hydrogenation was measured as a function of H.sub.2 and CO.sub.2 partial pressures at 500-600 K, 140 kPa and 30,000-90,000 h.sup.-1. The data indicated the rate of CO.sub.2 hydrogenation is moderately dependent on CO.sub.2 and H.sub.2 concentrations at low partial pressures but essentially concentration independent at high partial pressures. Under typical conditions CO was observed as a product of the reaction at levels determined by quasi-equilibrium between surface and gas phase CO species. Addition of CO to the reactants above this equilibrium level caused a significant decrease in the rate of CO.sub.2 hydrogenation apparently as a result of product inhibition.
These authors set forth the following table which in their view summarized the mechanism for CO.sub.2 hydrogenation and accounted for the observations of this study and other recent studies.
______________________________________
Proposed Sequence of Elementary Steps in CO.sub.2
Methanation.sup.a
Reaction
______________________________________
H.sub.2 (g) + S
##STR2## 2HS
CO(g) + 2S
##STR3## COS + OS
COS
##STR4## CO(g) + S
COS +
##STR5## CS + OS
CS + HS
##STR6## CHS + S
CHS + HS
##STR7## CH.sub.2S + S
CH.sub.2S + HS
##STR8## CH.sub.3S + S
CH.sub.3S + HS
##STR9## CH.sub.4S + S
CH.sub.4S
##STR10##
CH.sub.4 (g) + S
OS + HS
##STR11##
OHS + S
OHS + HS
##STR12##
H.sub.2 OS + S
H.sub.12 OS
##STR13##
H.sub.2 O(g) + S
______________________________________
.sup.a S refers to a surface site.
The authors also noted that kinetic studies of methanation in their laboratory provided evidence that there are several mechanistic regimes involving different rate controlling steps, the importance of which depends mainly on temperature.
The authors proposed that at moderate reaction temperatures (525.degree.-575.degree. K.), CO dissociation appeared to be rate controlling while hydrogenation of carbon apparently controlled at higher reaction temperatures (>575.degree. K.). This suggested that CO and CO.sub.2 methanation rates on nickel are both controlled by CO dissociation at moderate reaction temperatures (525-575.degree. K.), a hypothesis consistent with the observation of nearly identical specific reaction rates in CO.sub.2 and CO methanation on Ni/SiO.sub.2. The authors further proposed that it is possible for the two reactions to follow similar paths and still have different rate determining steps.
CO.sub.2 Methanation catalysts are discussed in an article titled "Highly Active Catalysts For CO.sub.2 Methanation To Provide The Second Reactor Of Two Stage Process For High BTU SNG Synthesis", A. D. Tomsett, et al., Applied Catal. 26 391 (1986). This investigation was made in the context of identifying highly active catalysts suitable for the second of two stages in a process for conversion of syngas to high BTU substituted natural gas. It was found that supported Ni-La.sub.2 O.sub.3 -Ru catalysts are effective for CO.sub.2 methanation. These catalysts provided high activity and complete conversion of CO.sub.2 to methane at as low a temperature as 250.degree. C.
It was concluded that the high activity of the Ni-La.sub.2 O.sub.3 -Ru catalysts is mainly ascribable to the large number of active sites.
It is known that methanation reactions are the reverse of reforming reactions. The methanation reactions are exothermic and, therefore, since a methanator typically operates in the temperature range of 300.degree.-400.degree. C. the CO and CO.sub.2 at the inlet should be carefully monitored to avoid damage to catalyst and vessel, see Kirk-Othmer, Encyclopedia of Chem. Tech., 3rd Edition, Vol. 2, p. 494.
Related art includes an investigation discussed in an article titled "Hydrogenation of CO.sub.2 over Co/Cu/K Catalysts", H. Baussart, et al., J. Chem. Soc. Faraday Trans. I, 83, 1711 (1987). Here Co, K and Cu, as components of catalysts were investigated for the specific roles each played in the hydrogenation of CO.sub.2. The results indicated, inter alia that a steady state was reached after ca. 30 hours of activation. It was found that the presence of cobalt favored the formation of alkanes, and that the presence of potassium favored the formation of CO resulting from a reverse water-gas shift reaction (CO.sub.2 +H.sub.2 .fwdarw.CO+H.sub.2 O). Further, it appeared the presence of copper in Russell catalysts increased the activity and the selectivity for CH.sub.4 and reduced the number of products.
There is a report on results of investigations into potassium-promoted nickel catalysts by T. K. Campbell, et al., Applied Catal., 50, 189 (1989). It was found potassium did not increase higher hydrocarbon or olefin selectivity, but it changed the CH.sub.4 /CO product distribution. It was also found that potassium did not change CO.sub.2 methanation mechanism and the mechanism appeared to be quite similar to that for CO hydrogenation. This paper presented steady-state kinetic data for CO.sub.2 hydrogenation on Ni/SiO.sub.2 and Ni/SiO.sub.2 -Al.sub.2 O.sub.3 catalysts for a range of potassium concentrations. It was reported that hydrogenation reactions of CO.sub.2 and CO on potassium promoted nickel catalysts are similar:
The activities and selectivities depend significantly on the support.
The rate of methanation on Ni/SiO.sub.2 decreases exponentially for both reactions with potassium addition.
The rate of methanation on Ni/SiO.sub.2 -Al.sub.2 O.sub.3 initially increases for both reactions with potassium addition.
Preoxidized rhodium was studied as a catalyst for methanation of carbon monoxide in an article titled "Methanation of Carbon Dioxide over Preoxidized Rhodium", A. Amariglio, et al., J. Catal., 81 247 (1983). The authors confirmed that according to their experiments preoxidation of Rh caused a dramatic enhancement of its activity in the methanation of CO.sub.2. They estimated preoxidation allowed the rate to be increased by a factor of 100 at the lowest estimate. The authors also concluded that the deactivation, observed upon prolonged exposures to H.sub.2, must be ascribed to the depletion of the preincorporated oxygen.
There is some information available regarding the use of ruthenium in a system structured for methanation alone. In an article entitled "Carbon Dioxide and Carbon Monoxide Methanation by Supported Ruthenium Catalysts", J. Chem. Soc. Chem. Commun., 955 (1982) by Helen Ferkul, et al., there is a discussion of ruthenium carbonyl clusters as active catalysts for the methanation of both carbon monoxide and carbon dioxide. The various carbonyl clusters were placed on different supports, such as .lambda.-alumina, silica gel or Na-Y zeolite. Added solvents were removed under reduced pressure and the catalyst systems containing 1% by weight of metal were activated by heating to 200.degree. C. in a stream of hydrogen. Catalytic activities were determined by passing synthesis gas through the catalysts in the temperature range of 200.degree.-400.degree. C. and space velocity of 1000-3750.sup.h-1. The major hydrocarbon product was methane. Though traces of C.sub.2 -C.sub.4 hydrocarbons were also obtained, the yields of C.sub.2 -C.sub.6 hydrocarbons apparently observed by others under similar conditions were not observed. (See T. Okuhara, et al., J. Chem. Soc., Chem. Commun., 1981, 1114). It was found that the best yields were obtained with Ru.sub.3 (CO).sub.12 on alumina. Comparison with the well-known RuCl.sub.3.3H.sub.2 O- Al.sub.2 O.sub.3 system showed that it did not deactivate as fast.
Additional information regarding the use of catalysts for CO.sub.2 methanation is found in an article by C. L. Thomas, entitled "Catalytic Processes and proven Catalysts", Academic Press., p. 116. It is noted that the most common ones are nickel. Many of the properties of the nickel catalysts are discussed along with disadvantages of using them in this type reaction. It is noted that in addition to nickel, platinum and ruthenium are used for methanation. There is no discussion here of the ruthenium catalyst.
From the art available it is evident that there are many factors and multiple variables involved in the methanation of CO.sub.2, such that it would be difficult to identify a catalyst which would promote methanation at low temperatures (<250.degree. C.) and high pH within a dual catalyst system where the conditions are geared toward a primary reaction, such as selective amination.
It would constitute a distinct advance in the art if a catalyst system were available whereby CO.sub.2 formation could be suppressed during morpholine/DGA amine processing by amination of DEG under anhydrous conditions. It has been discovered that not just any methanation catalyst could be used in this fashion because of certain critical factors, particularly those outlined supra.
Unexpectedly it has been discovered that a dual catalyst system comprising a nickel in conjunction with a supported ruthenium catalyst is very effective for selective amination of DEG to morpholine plus DIGLYCOLAMINE.RTM. amine while at the same time significantly suppressing the coproduction of CO.sub.2.
SUMMARY OF THE INVENTIONAccording to the broad aspect of this invention morpholine and DIGLYCOLAMINE.RTM. amine are simultaneously produced while at the sometime CO.sub.2 production is substantially suppressed by a process comprising contacting diethylene glycol, ammonia and hydrogen in the presence of a dual nickel/ruthenium catalyst comprising nickel-copper-chromium oxide in conjunction with a supported ruthenium catalyst at a temperature of 150.degree. C. to 300.degree. C. and pressure ranging from 200 to 5000 psi under anhydrous feed conditions, in the presence of a substantial quantity of hydrogen, and recovering the products.
DETAILED DESCRIPTION OF THE INVENTIONAccording to the preferred embodiment, oxydiethanol, also known as diethylene glycol, is fed into a reactor under anhydrous conditions and contacted with ammonia and hydrogen in the presence of a catalytically effective amount of a dual catalyst system comprising an amination/hydrogenation/dehydrogenation catalyst and a methanation catalyst, wherein the amination/hydrogenation/dehydrogenation catalyst consists of an oxide of 60 to 85 mole % nickel, about 14 to 37 mole % copper and 1 to 6 mole % chromium, and the methanation catalyst comprises a supported ruthenium catalyst.
The reaction takes place in a continuous process and contacting of the reactants is accomplished at a temperature of from 150.degree. C. to about 300.degree. C. and at a pressure of about 200 psig to about 5000 psig. The molar ratio of ammonia to oxydiethanol is preferably from about 4:1 to about 8:1 and the space velocities of the liquid feed material, i.e. ammonia and oxydiethanol, are from about 1 g of liquid feed per hour per milliliter of catalyst to about 6 g of liquid feed per hour per milliliter of catalyst. The recovered reaction product, consisting primarily of DIGLYCOLAMINE.RTM. amine and morpholine, is recovered from the resultant crude reaction product by conventional means such as distillation, extraction and the like.
The oxydialkanol compound that can be utilized in practicing the instant invention can be of the general formula: ##STR14## wherein each R is, independently, a hydrogen or a lower alkyl radical, such as, for example, methyl, ethyl or the like. The examples demonstrate the effectiveness of oxydiethanol or diethylene glycol.
The catalyst for amination of diethylene glycol which may be employed in practicing the invention may generally be characterized as an amination catalyst or as an amination/hydrogenation catalyst. Examples include copper, nickel, chromium, cobalt, manganese, molybdenum, palladium, platinum, rhodium, oxides of such metals and mixtures thereof. A particularly preferred catalyst is characterized by having the composition calculated in mole % of from about 60-85% nickel, 14-37% copper and 1-6% chromium with the preferred proportions being about 72-78% nickel, 20-25% copper and 1-3% chromium. A greatly preferred hydrogenation catalyst is one in which the active components consist essentially of about 75 mole % nickel, about 23 mole % copper and about 2 mole % chromium in the form of an oxide.
A commercially available amination/hydrogenation catalyst fitting this description is Engelhard Ni-2715 T 3/16". Engelhard Ni-2715 is a nickel-based catalyst produced by Engelhard which contains a nickel-copper-chromium oxide.
As mentioned only certain combinations of nickel-ruthenium based dual catalyst systems are effective in suppressing CO.sub.2 formation during morpholine service and it has been surprisingly discovered in the instant invention that an effective methanation catalyst in combination with the nickel-copper-chromium oxide catalyst, is a specified class of supported ruthenium catalysts. Said supports may be selected from the elements of the Group II, III and IV or oxides thereof. The preferred supports include carbon, magnesia, alumina, silica, zirconia and titania, as well as mixtures thereof. The most preferred support is carbon.
The ruthenium may be added to said support by any of the usual methods and said formulated catalyst may comprise from 0.01 to 30 wt % ruthenium. The preferred ruthenium loading is 0.1 to 10.0 wt %, and the preferred catalyst formulation comprises 0.5 to 5.0 wt % ruthenium-on-carbon. Suitable commercially-available catalysts include Engelhard's 0.5% ruthenium-on-granular carbon. Where the support is alumina, the preferred ruthenium loading is 0.1 to 10 wt %.
The order of activity of the different ruthenium cocatalysts appears to be 1% Ru/C.about.0.5% Ru/C>5% Ru/Al.sub.2 O.sub.3 .apprxeq.0.5 Ru/Al.sub.2 O.sub.3.
Said combination of nickel/ruthenium dual catalysts may be used in any proportions except that preferably the nickel-copper-chromium oxide catalyst should contact the diethylene glycol/ammonia feed first, in order to achieve the desired amination (Equations 1 and 2), then after the desired DEG conversion is achieved, the reactant product should contact the supported ruthenium catalyst in order to significantly lower the CO.sub.2 effluent concentration (through Reaction 3).
A particularly effective ratio of nickel-copper-chromium oxide catalyst to supported ruthenium catalyst is in the range 1:1 to 10:1 by volume. The accompanying examples illustrate such a ratio range. A particularly preferred dual nickel catalyst combination is Engelhard Ni-2715 T 3/16" and 0.5 wt % ruthenium-on-carbon.
The ratio of reactant, i.e. the ratio of ammonia to oxydiethanol, or other oxydialkanol used in this process is 1:1 to about 10:1 although a molar excess of ammonia produces increased yields of the morpholine product. Optimum yields are obtained when the molar ratio of ammonia to oxydiethanol is about 6:1.
The space velocities of the feed, namely, the ammonia and the glycol, may vary from about 1 gram of the liquid feed/hour/milliliter of catalyst, to about 6 grams of liquid feed/hour/milliliter of catalyst. As noted above, the temperature range for the present invention is between 150.degree. C. and 300.degree. C., the preferred range is 200.degree. to 250.degree. C., depending upon the reactant ratios utilized.
Those skilled in the art will realize that an outstanding advantage of the instant process resides in the fact that the relative yields of morpholine and DIGLYCOLAMINE.RTM. amine can be varied by a slight variation of reaction conditions while the production rate of the combined products remains high. Thus, one can achieve a substantial yield of 2-(2-aminoethoxy)ethanol while simultaneously maintaining morpholine production rates comparable to or better than those obtained by previously known methods. In addition, no substantial by-products are formed.
The process should be carried out in a reductive atmosphere using hydrogen for commercial operations. Although small amounts of hydrogen are present during the reaction from the catalyst materials and as a by-product, it is preferred that hydrogen be added in the feed system in order of achieved CO.sub.2 methanation (Equation 3) and to maintain catalytic activity. The quantity of hydrogen necessary for the practice of this invention is generally in the range of from 0.1 to 1000 liters (1) per lb of DEG converted, the preferred range is from 10 to 200 liters H.sub.2 per lb of DEG converted. Such a ratio is illustrated in the accompanying examples. Data for the combination of Engelhard Ni-2715T and ruthenium-on-carbon show CO.sub.2 production may be suppressed by a factor of >2 where the DEG conversion is ca. 80% and by a factor of ca. 5, when the H.sub.2 flow rate is raised from 35 liters to 90 liters/hr.
The quantity of CO.sub.2 produced in this morpholine/DIGLYCOLAMINE.RTM. amine process is normally measured in terms of parts per million (ppm) in the effluent product and is expressed in terms of lbs of CO.sub.2 generated per 100M lb of DEG converted.
The accompanying examples illustrate the following:
1) In Example I, the generation of morpholine plus DIGLYCOLAMINE.RTM. (DGA.RTM. amine) is demonstrated using a dual catalyst system comprising a nickel-copper-chromium oxide catalyst in combination with a ruthenium-on-carbon to achieve:
a) At normal hydrogen flows, 81% DEG conversion at 215.degree. C. with 248 lb CO.sub.2 produced per 100M lb. DEG converted (see Table I).
b) At 2.6 times normal hydrogen flows, ca 85% DEG conversion at 215.degree. C. with 93-109 lb CO.sub.2 produced per 100M lb DEG converted.
2) In Examples 2 and 3, similar combinations with other ruthenium-on-carbon catalysts typically give:
a) At normal hydrogen flow rates, 80% DEG conversion at 215.degree. C. with 224 lb CO.sub.2 produced per 100M lb DEG converted (see Table III).
b) 2.6 times normal hydrogen flows, on 90% DEG conversion at 220.degree. C. with 88-102 lb CO.sub.2 produced per 100M lb DEG converted.
3) In Examples 4 and 5, combinations of the Ni-Cu-Cr catalyst with ruthenium-on-alumina catalyst typically give:
a) At normal hydrogen flow rates, 75% DEG conversion at 210.degree. C. with 251 lb CO.sub.2 produced per 100M lb DEG converted (see Table IV).
b) At 2-6 times normal hydrogen flows, ca. 88% DEG conversion at 220.degree. C. with 226-231 lb CO.sub.2 produced per 100M lb DEG converted.
4) By comparison, in comparative Example A, using only Ni-2715 T 3/16", when DEG is aminated at 220.degree. C. with 82% DEG conversion, production of by-product CO.sub.2 is 499 lb per M lb DEG converted.
5) In Example 6, combinations of the Ni-Cu-Cr catalyst with ruthenium-on-carbon catalyst provide improved CO.sub.2 suppression during a typical 500 hr experiment.
EXAMPLE 1To a 55 cc plug-flow reactor fitted with heating, cooling and pressure control devices and constructed of stainless steel, was charged a combination of 400 cc of pelleted nickel-copper-chromium oxide catalyst (in which the active components consisted essentially of 71.2 wt % Ni, 12.3 wt % Cu and 73.3% reduced Ni) and 150 cc of granular ruthenium (0.5%) on carbon, 4-12 mesh. The pelleted nickel-copper-chromium oxide catalyst was placed in the bottom section of the reactor and the ruthenium-on-carbon at the top.
Diethyleneglycol (1.2 lbs/hr), ammonia (2.4 lbs/hr) and hydrogen (35 liter/hr) were fed separately to the reactor system from the bottom of the reactor, in an upflow mode, through the catalyst bed of the reactor. The reactor effluent was cooled, depressurized and the gaseous and liquid product analyzed after a 48 hour prerun period to establish steady state conditions, the performance of the dual-catalyst system was tested at a series of operating temperatures (210.degree.-230.degree. C.) and hydrogen flow rates (35-90 liters/hr) while maintaining a back pressure of 2200 psig.
Data and results from this run are summarized in Table I. It may be noted that in this run:
At an operating temperature of 215.degree. C., the DEG conversion is 80.6% and the CO.sub.2 production is 248 lbs per 100M lb of DEG converted.
While at 220.degree. C. temperature, when the hydrogen flow rate is raised to 90 liters/hr, the CO.sub.2 effluent production drops to 93-109 lbs per 100M lb of DEG converted.
TABLE I
__________________________________________________________________________
CO.sub.2 (lbs)
DEG Effluent
Per
H.sub.2 Flow
Reaction
Conver-
Yield (%) Basis
CO.sub.2
100 M lbs
Rate Temp.
sion DEG Converted
Conc.
DEG Con-
EX.
(l/hr)
(.degree.C.)
(%) Morpholine
DGA (ppm)
verted
__________________________________________________________________________
1.sup.a
35 210 64.5 49.0 45.2
.sup.b
.sup.b
35 215 80.6 69.6 25.6
190 248
35 220 86.9 75.5 18.8
305 357
35 230 95.0 84.3 9.3 790 856
90 220 84 68 15 92 109
90 220 85.9 74 15 80 93
__________________________________________________________________________
.sup.a Catalyst combination: Ni--Cu--Cr (400 cc) + 0.5% Ru/C, 4-12 Mesh
(150 cc).
.sup.b Not determined.
EXAMPLES 2-3
Following the operating procedures of Example 1, two combinations of the nickel-copper-chromium oxide catalyst of Example 1 with different ruthenium-on-carbon catalysts (A 0.5% Ru/C, 30.times.40 mesh, and a 1% ruthenium-on-carbon, granules) were tested for diethylene glycol amination over a range of operating temperatures (210.degree.-230.degree. C.) and the different hydrogen flow rates (35 and 90 liters/hr).
Data and results from these runs are summarized in Tables II and III. It may be noted that:
a) At normal hydrogen flow rates, 80% DEG conversion per pass is achieved at 215.degree. C., with 224 lb CO.sub.2 produced for each 100M lb DEG converted (see Table III).
b) At 2.6 times the normal hydrogen flows, ca. 90% DEG conversion is achieved at 220.degree. C., with 88-102 lb CO.sub.2 produced per 100M lb DEG converted.
TABLE II
__________________________________________________________________________
CO.sub.2 (lbs)
DEG Effluent
Per
H.sub.2 Flow
Reaction
Conver-
Yield (%) Basis
CO.sub.2
100 M lbs
Rate Temp.
sion DEG Converted
Conc.
DEG Con-
EX.
(l/hr)
(.degree.C.)
(%) Morpholine
DGA (ppm)
verted
__________________________________________________________________________
2.sup.a
35 210 67.3 56.1 39.4
112 162
35 215 73.7 62.1 33.1
136 152
35 220 82.5 70.1 24.8
236 237
35 230 91.4 79.7 13.3
790 832
90 220 79.8 72.7 19.9
97 124
90 220 80.8 71.2 19.8
110 140
__________________________________________________________________________
.sup.a Catalyst combination: Ni--Cu--Cr (400 cc) + 0.5% Ru/C, 30 .times.
40 Mesh (150 cc).
TABLE III
__________________________________________________________________________
CO.sub.2 (lbs)
DEG Effluent
Per
H.sub.2 Flow
Reaction
Conver-
Yield (%) Basis
CO.sub.2
100 M lbs
Rate Temp.
sion DEG Converted
Conc.
DEG Con-
EX.
(l/hr)
(.degree.C.)
(%) Morpholine
DGA (ppm)
verted
__________________________________________________________________________
3.sup.a
35 210 72.1 61.2 34.1
102 153
35 215 80.4 68.4 26.6
167 224
35 220 85.2 73.4 21.6
282 344
35 230 95.6 85.2 8.8 931 992
90 220 90.4 77.3 12.9
75 88
90 220 90.4 78.8 11.8
87 102
__________________________________________________________________________
.sup.a Catalyst combination: Ni--Cu--Cr (400 cc) + 1% Ru/C, granules (50
cc).
EXAMPLES 4-5
Following the operating procedures of Example 1, two combinations of the nickel-copper-chromium oxide catalyst of Example 1 with different ruthenium-on-alumina catalysts (a 5% ruthenium-on-alumina, 1/8" diameter extrudates and 0.5% ruthenium-on-alumina, 1/8" diameter spheres) were tested for diethylene glycol amination over a range of operating temperatures (210.degree.-250.degree. C.) and, in one case, different hydrogen flow rates (35 and 90 liters/hr).
Data and results from these runs are summarized in Table IV and V. It may be noted that typically:
a) At normal hydrogen flows, 75% DEG conversion per pass is achieved at 210.degree. C., with 251 lb of CO.sub.2 produced per 100M lb DEG converted (see Table IV).
b) At 2.6 times the normal hydrogen flows, ca. 88% DEG conversion is achieved at 220.degree. C., with 226-231 lb CO.sub.2 produced per 100M lb DEG converted.
TABLE IV
__________________________________________________________________________
CO.sub.2 (lbs)
DEG Effluent
Per
H.sub.2 Flow
Reaction
Conver-
Yield (%) Basis
CO.sub.2
100 M lbs
Rate Temp.
sion DEG Converted
Conc.
DEG Con-
EX.
(l/hr)
(.degree.C.)
(%) Morpholine
DGA (ppm)
verted
__________________________________________________________________________
4.sup.a
35 210 74.7 63.0 31.1
179 251
35 215 .sup.b
.sup.b
.sup.b
279 393
35 220 84.9 73.3 21.4
440 543
35 230 94.5 84.1 9.8 1241 1375
90 220 88.4 77.6 13.0
212 226
90 220 87.1 75.9 14.7
178 231
__________________________________________________________________________
.sup.a Catalyst combination: Ni--Cu--Cr (400 cc) + 5% Ru/alumina, E1/8"
(150 cc).
.sup.b Not determined
TABLE V
__________________________________________________________________________
CO.sub.2 (lbs)
DEG Effluent
Per
H.sub.2 Flow
Reaction
Conver-
Yield (%) Basis
CO.sub.2
100 M lbs
Rate Temp.
sion DEG Converted
Conc.
DEG Con-
EX.
(l/hr)
(.degree.C.)
(%) Morpholine
DGA (ppm)
verted
__________________________________________________________________________
5.sup.a
35 210 52.5 45.7 51.4
90 182
35 220 83.0 75.6 20.7
330 419
35 230 93.0 83.0 11.1
510 531
35 240 94.5 78.0 9.8 2770 3060
35 250 96.7 80.4 7.3 3190 3534
__________________________________________________________________________
.sup.a Catalyst combination: Ni--Cu--Cr (400 cc) + 0.5% Ru/Alumina, S1/8"
(150 cc).
EXAMPLE 6
Following the procedures of Example 1, the combination of nickel-copper-chromium oxide catalyst with 0.5% ruthenium-on-carbon, granular, catalyst was tested for diethylene glycol amination over a period of 500 hrs. Data and results from this run are summarized in Table VI. It may be noted that:
1) The dual Ni-Ru catalyst continues to suppress CO.sub.2 effluent concentrations and actually becomes more effective as the run progresses. After 400 hrs, the CO.sub.2 effluent concentrations are below normal threshold levels.
2) The DEG conversion level remains high--from ca. 88% to 80%--over the life of the catalyst.
TABLE VI
__________________________________________________________________________
CO.sub.2 (lb)
per
Time on DEG conv.
Product yield, %
CO.sub.2 conc.
100 M lb
EX.
stream (hrs)
% Morpholine
DGA (ppm) DEG conv.
__________________________________________________________________________
6.sup.a
22 88.7 78.8 14.6
189 233
70 87.9 79.2 14.5
126 151
100 85.6 75.5 17.9
125 167
171 83.5 73.6 20.1
112 143
203 82.1 72.6 21.5
165 210
203 82.1 72.6 21.5
165 210
273 81.8 72.5 22.0
108 146
303 83.1 73.8 20.0
116 153
375 82.5 72.9 20.9
126 155
401 80.2 68.7 23.9
29 37
449 80.8 70.2 23.6
31 40
484 79.7 69.0 24.9
24 31
__________________________________________________________________________
.sup.a Catalyst combination: Ni--Cu--Cr (400 cc) + 0.5% Ru/C, 4-12 mesh
(150 cc).
COMPARATIVE EXAMPLE A
Following the operating procedures of Example 1, the 550 cc plug flow reactor was filled with the nickel-copper-chromium oxide catalyst alone and diethylene glycol amination was conducted at feed rates of 1.2 lb/hr DEG, 2.4 lb/hr of ammonia and 35 liters/hr of hydrogen. Again the operating temperature and pressure were 220.degree. C. and 220 psi respectively.
Analyses of the liquid and gas products show:
DEG Conversion - 82%
Yield of Morpholine - 65.3%
Yield Of DGA - 30.3%
Effluent Conc. CO.sub.2 - 689 ppm
CO.sub.2 Per 100M Lbs DEG Converted - 499 lbs
Claims
1. A process for simultaneously producing morpholine and DIGLYCOLAMINE.RTM. amine while suppressing CO.sub.2 formation comprising the steps of:
- reacting anhydrous diethylene glycol with ammonia in the presence of added hydrogen over a dual nickel/ruthenium catalyst system consisting essentially of a nickel-copper-chromium oxide catalyst in combination with a supported ruthenium catalyst at a temperature of from about 150.degree. C. to about 300.degree. C. and a pressure of from about 200 psig to about 5000 psig.
2. The process of claim 1 wherein the ruthenium catalyst comprises 0.1 to 10 ruthenium.
3. The process of claim 2 wherein the ruthenium catalyst support is selected from the group comprising carbon, alumina, silica, titania, zirconia, and magnesia, as well as mixtures thereof.
4. The process of claim 3 wherein the support is carbon.
5. The process of claim 1 wherein the nickel-copper-chromium oxide component of the dual catalyst comprises:
- 60 to 85 mole percent nickel,
- 14 to 37 mole percent copper,
- 1to 6 mole percent chromium
6. The process of claim 1 wherein the hydrogen flow is from 10 to 200 liters per lb of DEG converted.
7. The process of claim 4 wherein the ruthenium-on-carbon catalyst comprises 0.5 to 5.0 ruthenium.
8. In a process for simultaneously producing morpholine plus DIGLYCOLAMINE.RTM. amine, the improvement of suppressing carbon dioxide formation by:
- reacting anhydrous diethylene glycol with ammonia in the presence of added hydrogen over a dual nickel/ruthenium catalyst which consists essentially of a nickel-copper-chromium oxide catalyst comprising 72-78% nickel, 20-25% copper and 1-3% chromium in combination with a supported ruthenium catalyst consisting essentially of 0.5-5.0% ruthenium on a support selected from the group consisting of carbon or alumina, at a pressure of 200 psi to 5000 psi and a temperature from 200.degree. to 250.degree. C., at hydrogen flow rates of from about 10 to 200 liters per lb. of DEG converted and isolating said morpholine and DIGLYCOLAMINE.RTM. amine products from the reaction products.
9. The process of claim 8 wherein the ratio of nickel-copper-chromium oxide catalyst to ruthenium-based catalyst, is in the range of 1:1 to 10:1 by volume.
| 4508896 | April 2, 1985 | Templeton |
Type: Grant
Filed: Dec 21, 1990
Date of Patent: Dec 10, 1991
Assignee: Texaco Chemical Company (White Plains, NY)
Inventor: John F. Knifton (Austin, TX)
Primary Examiner: Robert W. Ramsuer
Attorneys: Jack H. Park, Kenneth R. Priem, Cynthia L. Kendrick
Application Number: 7/632,205
International Classification: C07D295023;
